LAKE D E S T R A T I F I C A T I O N A N D S P E C I A T I O N OF I R O N A N D MANGANESE B A R R Y C H I S W E L L * and M Y I N T Z A W
Department of Chemistry, University of Queensland, St. Lucia, Qld, Australia 4067
(Received March 1991) Abstract. A long-term study of the effect of artificial aeration (destratification) of a water storage dam upon the speciation of iron and manganese in the d a m waters has been undertaken. Separation of d a m samples into soluble and insoluble forms by selective membrane filtration was undertaken before using the techniques of EPR spectroscopy, ion chromatography and gel filtration to assess the speciation of soluble species, and selective extraction and surface analysis (ESCA, SIMS and SEM) techniques to determine the speciation of particulate iron and manganese species. The percentages of soluble iron and manganese before (1983-85) and after (1986-88) artificial aeration are compared for the periods Jan-Dec, Jan-Mar, and J u n - A u g at three depths 6 m, 15 m and 0.5 m above the d a m base, to assess the importance of seasonal changes in the various depths of the dam. Although aeration had an initial marked reduction in levels of soluble iron and manganese at all depths of the dam, the concentrations of these totals showed a steady increase over succeeding years. Analysis of the figures over summer and winter periods shows that the reduction of soluble iron was maintained in summer, but not during winter. Upon aeration, the initial reduction of soluble manganese concentration was maintained in succeeding years in the epilimnic regions of the dam, but not in the hypolimnion. Statistical analysis of data has been undertaken to correlate the changes in relationship between the various forms of iron and manganese with the advent of aeration.
Introduction M A N G A N E S E A N D IRON IN POTABLE WATERS
Both manganese and iron at very low concentrations are highly objectionable in water supplies for domestic and industrial use. Trace amounts of iron and manganese can cause staining of bathroom fixtures, equipment, manufactured goods and swimming pools, can impart brownish colour to laundered clothing, and can affect the taste of water. The concentrations of these elements in drinking waters often exceeds the National Health and Medical Research Council recommended values of 0.1 mg 1-l for iron and 0.05 mg 1-1for manganese (NH and MRC, 1980) with associated increases in consumer complaints. The presence of iron and manganese in piped water supply systems (including urban reticulation and hydro-electric systems) has been shown to create some significant problems (Loos, 1987), which arise when, either microbioorganisms present in the water metabolise the iron and manganese depositing insoluble iron and manganese oxides along the growing bacterial stalk of the organism, or during the purification of drinking water by chlorination, deposition of brown iron and black manganese oxides from solution occurs. The growth of bacterial deposits can reduce the volume of pipelines and reduce the flow rate of water in both town water supply systems (Faust and Aly, 1983; WHO) and * Author to whom correspondence should be addressed.
Environmental Monitoring and Assessment 19: 433-447, 1991. 9 1991 Kluwer Academic Publishers. Printed in the Netherlands.
434
BARRY CHISWELL AND MYINT ZAW
hydroelectric (Tyler and Marshall, 1967; Tyler, 1970; Marshall, 1980) systems. Both the microbiologically produced iron and manganese oxides, and the chlorine-precipitated iron and manganese oxides yield dirty brown-black water which is the cause of consumer complaints. LAKE STRUCTURE AND STRATIFICATION
Deeper lakes in temperate regions are generally subject to gradual heating of the waters in the spring and summer resulting in a characteristic form of thermal stratification (Coker et al., 1979; Timperley and Allan, 1974) consisting of the epilimnion, and upper region of generally uniformly warm, circulating and fairly turbulent water because of the effects of wind action and convection currents; (ii) the hypolimnion, a bottom layer of cold and relatively undisturbed water, which is isolated from wind and thermal effects and is generally devoid of light and dissolved oxygen, separated from the epilimnion by, (iii) the metalimnion, and intermediate region in which the temperature gradient, the thermocline, is steepest. (i)
The term thermocline has been defined variously, but correctly refers to the plane or surface of maximum rate of decrease of temperature with depth (Wetzel, 1975; Chiswell and Rauchle, 1986). The metalimnion in terms of an oxycline rather than thermocline has been used in relation to destratification and speciation of iron and manganese. The oxycline, over which there is a sudden change in dissolved oxygen (D.O.) values, may often coincide with the thermocline but is usually much more clearly cut. There are many different forms of stratification brought about by the form, size, depth and location of the lake basin, volume of through-flow, and effects of climate (Coker et al., 1979). With the onset of thermal stratification, deep circulation within the water column ceases and the oxygen transport from the productive zones near the surface to the deeper part of the water column is restricted. Bacterial consumption of oxygen does not cease following stratification, and in the absence of replacement from the surface zone the water/sediment interface rapidly becomes anaerobic and the overall oxygen content of the hypolimnion drops rapidly. The anaerobic production of carbon dioxide (Hutchinson, 1975), in conjunction with the lack of photosynthetic activity within the oxygen deficient region, usually results in the lowering of pH. Decomposition of settling organic material within the anoxic hypolimnion and at the water/sediment interface also leads to the release of hydrogen sulfide and ammonia. The development of chemically reducing conditions at the water/sediment interface leads to the release of iron and manganese compounds from the sediments into the water body (Uhlmann, 1979; Davison and Woof, 1984). Dissolved iron and manganese may adsorb onto biogenic and organically coated particles (Ballstrieri et al., 1981; Martin and Knauer, 1985), or oxidize during the formation of oxyhydroxide coatings on bacterial capsules (Cowen and Silver, 1984; Cowen and Bruland, 1985). Iron and manganese may be incorporated into organic tissue
LAKE
DESTRATIFICATION
AND
SPECIATION
OF FE AND
MN
435
as necessary biochemical micronutrients, perhaps to be redissolved upon the recycling of these m~iterials with depth. Thus biological stratification can also influence the distribution of iron and manganese in a water body. ARTIFICIAL DESTRATIFICATIONOF DAM WATER Aeration/destratification is a management technique which has been widely used internationally and in Australia to effect beneficial changes in the thermal, chemical and biochemical status of both man-made and naturally occurring water storage. The method is now recognized as an effective way of minimizing the release of iron and manganese from sediments and is being used increasingly in Australian reservoirs for water quality improvement purposes (Loos 1987). Oxygen saturation maintained at above 5% throughout the water column is sufficient to prevent high concentrations of iron and manganese from developing (Bowles et al. 1979). Two additional benefits of the introduction of artificial destratification are the maintenance of uniform water quality within the reservoir throughout the year, and the maintenance of maximum residence time of soluble metal ions within the reservoir. This must be done to ensure complete formation of particulate forms of iron and manganese with subsequent sedimentation of iron and manganese oxides from the water column (Waite et al., 1989).
Aims of the Research The work described here was undertaken with a view to developing a deeper understanding of the forms of occurence (speciation) of manganese and iron in impounded waters such as storage dams. Within this deeper understanding, there was seen the need to elucidate how speciation of manganese and iron changed, and what changes in naturally occuring parameters led to such changes. The work had a particular applied aim in seeking to use the knowledge obtained about speciation to develop both the understanding, and the application, of methods used to remove these nuisance metals from potable water during water treatment. Because studies were also carried out on an artifically destratified dam, new insights into the effectiveness of aeration in dealing with manganese and iron removal were also able to be studied.
Methodology Employed Two major techniques of manganese and iron speciation were developed for the first time for use on darn water storages. These were the use of electron paramagnetic spectroscopy (EPRS) to speciate dissolved manganese, and the use of secondary ion mass spectroscopy (SIMS) and electron spectroscopy for chemical analysis (ESCA) to speciate particulate manganese and iron. Other techniques used have previously been employed in a similar manner to the use made in the work here described.
436
BARRY CHISWELL AND MYINT ZAW
DISSOLVED AND PARTICULATESPECIES For the purposes of this work, the distinction between these two forms of a metal is related to passage through (dissolved) and retention by (particulate) a filter membrane; usually of pore size 0.45 #m, but sometimes 0.01 #m. SPECIATION OF DISSOLVEDMN AND FE
EPR Spectroscopy of Dissolved Mn This work has developed the use of EPRS to yield the analytical determination of aquated manganese(II) in freshwaters. The technique has been shown to be very sensitive (detection limit of 5 #g 1-1)and to be as accurate as flame atomic adsorption spectroscopy for manganese; it also has the great advantage of being selective for aquated manganese(II) only.
Gel Filtration~Molecular Weight Speciation of Mn and Fe The method of gel (or particle size exclusion) filtration has been applied to the waters from both North Pine and Hinze dams. The object of this work is to speciate manganese and iron dissolved species by molecular weight ranges. The gels used, Sephadex, G-10 and G-25, have approximate molecular weight operational ranges of 1 to 700 and 1 to 5000 Daltons respectively; in theory compounds within these molecular weight ranges are held on the chromatography column, while compounds with molecular weights above the upper limit of the range are excluded from the gel and pass out of the column. This work has been coupled with the use of the ion-exchange column material, Chelex100, which is a strong binder of cationic species, but allows the passage of neutral organic complexes.
Membrane Filtration and Speciation of Mn and Fe The use of both 0.45 #m and 0.01 #m membrane filters upon like samples has been applied to give further evidence upon molecular/particle size of species in samples.
Speciation of Dissolved Fe(II) and Fe(II1) This has been achieved by a variation of the colorimetric analytical method which uses the dye ferrozine. This material, which is ion-specific for Fe(II), can be used to first determine Fe(II) and then, after a reduction step upon the sample, can also determine the Fe(III) originally present, this analysis has been used in particular to estimate the levels of complexation of Fe(II) and Fe(III), by coupling this work with gel filtration results. SPECIATION OF PARTICULATE
MN AND FE
Surface and Bulk Analysis of Material by SIMS and ESCA Coupled with a detailed study of particulates, taken from different depths of a dam and during different seasons, by scanning electron microscopy (SEM), surface analysis techniques have been used to analyse the oxidation state and binding atoms for Mn and Fe
LAKE
DESTRATIFICATION
AND
SPECIATION
OF
FE
AND
MN
437
present both on the surface and in the bulk of the particulates. The technique, although not previously applied to such a study, yields excellent information upon the nature of the particulate surface, and then, by ion-bombardment to remove approximately the top ten atomic layers of the solid, the nature of the bulk surface can be studied.
Sequential Extraction and Differential Solubility Studies of Mn and Fe in Particulates The nature of binding of both metals in particulate samples has been studied by these two closely related techniques. Sequential extraction of samples consists of step-wise extraction using increasingly more vigorous reactants; a typical program would have the sequence: (i) (ii) (iii) (iv) (iv)
equilibration with magnesium chloride sodium acetate/acetic acid extraction hydroxylamine/acetic acid extraction hydrogen peroxide/acetic acid extraction hydrofluoric/perchloric acids extraction in an attempt to define the amounts of Mn and Fe present as: (i) absorbed ionic material, (ii) carbonate-bound metals, (iii) oxide-bound metals, (iv) organically-bound metals, (v) residual (probably in clays) metals, respectively.
Differential solubility in various acids has been coupled with this work. DAM WATERSSTUDIEDAND SAMPLINGPROGRAMSUSED Two dams in South East Queensland have been extensively studied by most of the methods mentioned above. These are North Pine dam, approximately 50 km north west of Brisbane, and Hinze dam, approximately 100 km south west of Brisbane. Both dams are major storages; North Pine has a capacity of 202 000 M1 and supplies the northern suburbs of Brisbane, while Hinze dam has a capacity of 164 000 M1 and serves most of the rapidly growing coastal area south of Brisbane. A third, small dam Lake Kurwongbah, which is very close to North Pine dam has also been studied in far less detail. Work since 1986 has involved regular (usually fortnightly - but sometimes weekly) sampling of the dams by research students and the leader of the Water Research Group. Such sampling has often been coupled with sampling programs undertaken by the local Council authorities, however, it has been a rule of sampling that each worker collects and pre-treats his/her own samples. Sampling for depth samples was undertaken by use of double-ended inlet-outlet depth sampler of various (including a 'home-made' version constructed in the Chemistry department workshop) makes, while sediment samples were obtained by a grab sampler which penetrated approximately 5 cm into the sediment. Attempts to sample top layers of the sediment were not successful; an electronic eye system which supposedly measured
438
BARRY
CHISWELL
AND
MYINT
ZAW
when a micro-sampler was entering the sediment, worked within a few metres of the water surface. In all three dams the algal population is always such that one cannot see the bottom of the dam at even a depth of a metre. Samples once collected were stored in polythene bottles, which had been thoroughly cleaned by repeated acid washing followed by distilled water cleansing. Filtration of samples was often done immediately following sampling, although this was not possible when using 0.01/~m membranes. Parameters such as pH, dissolved oxygen, temperature and conductivity where either measured in situ or immediately following collection of the sample. Results and Discussion
SPECIATION OF DISSOLVEDMANGANESE The method of electron paramagnetic resonance spectroscopy (EPRS) (Chiswell and Mokhtar, 1986 and 1987) has been employed upon waters obtained from the epilimnion, the hypolimnion and the sediment of the three different dams to show: (i) that waters taken from any depth in the dam, and filtered through a 0.01 #m membrane contain dissolved manganese in the form of aquated Mn(II) only; such filtrate samples have identical concentrations of Mn(II) aq as determined by EPRS and atomic absorption spectroscopy (AAS) (see Table I). (ii) that waters filtered through a 0.45 ~t membrane from the epilimnic regions of the dam may contain a small concentration (
Department of Chemistry, University of Queensland, St. Lucia, Qld, Australia 4067
(Received March 1991) Abstract. A long-term study of the effect of artificial aeration (destratification) of a water storage dam upon the speciation of iron and manganese in the d a m waters has been undertaken. Separation of d a m samples into soluble and insoluble forms by selective membrane filtration was undertaken before using the techniques of EPR spectroscopy, ion chromatography and gel filtration to assess the speciation of soluble species, and selective extraction and surface analysis (ESCA, SIMS and SEM) techniques to determine the speciation of particulate iron and manganese species. The percentages of soluble iron and manganese before (1983-85) and after (1986-88) artificial aeration are compared for the periods Jan-Dec, Jan-Mar, and J u n - A u g at three depths 6 m, 15 m and 0.5 m above the d a m base, to assess the importance of seasonal changes in the various depths of the dam. Although aeration had an initial marked reduction in levels of soluble iron and manganese at all depths of the dam, the concentrations of these totals showed a steady increase over succeeding years. Analysis of the figures over summer and winter periods shows that the reduction of soluble iron was maintained in summer, but not during winter. Upon aeration, the initial reduction of soluble manganese concentration was maintained in succeeding years in the epilimnic regions of the dam, but not in the hypolimnion. Statistical analysis of data has been undertaken to correlate the changes in relationship between the various forms of iron and manganese with the advent of aeration.
Introduction M A N G A N E S E A N D IRON IN POTABLE WATERS
Both manganese and iron at very low concentrations are highly objectionable in water supplies for domestic and industrial use. Trace amounts of iron and manganese can cause staining of bathroom fixtures, equipment, manufactured goods and swimming pools, can impart brownish colour to laundered clothing, and can affect the taste of water. The concentrations of these elements in drinking waters often exceeds the National Health and Medical Research Council recommended values of 0.1 mg 1-l for iron and 0.05 mg 1-1for manganese (NH and MRC, 1980) with associated increases in consumer complaints. The presence of iron and manganese in piped water supply systems (including urban reticulation and hydro-electric systems) has been shown to create some significant problems (Loos, 1987), which arise when, either microbioorganisms present in the water metabolise the iron and manganese depositing insoluble iron and manganese oxides along the growing bacterial stalk of the organism, or during the purification of drinking water by chlorination, deposition of brown iron and black manganese oxides from solution occurs. The growth of bacterial deposits can reduce the volume of pipelines and reduce the flow rate of water in both town water supply systems (Faust and Aly, 1983; WHO) and * Author to whom correspondence should be addressed.
Environmental Monitoring and Assessment 19: 433-447, 1991. 9 1991 Kluwer Academic Publishers. Printed in the Netherlands.
434
BARRY CHISWELL AND MYINT ZAW
hydroelectric (Tyler and Marshall, 1967; Tyler, 1970; Marshall, 1980) systems. Both the microbiologically produced iron and manganese oxides, and the chlorine-precipitated iron and manganese oxides yield dirty brown-black water which is the cause of consumer complaints. LAKE STRUCTURE AND STRATIFICATION
Deeper lakes in temperate regions are generally subject to gradual heating of the waters in the spring and summer resulting in a characteristic form of thermal stratification (Coker et al., 1979; Timperley and Allan, 1974) consisting of the epilimnion, and upper region of generally uniformly warm, circulating and fairly turbulent water because of the effects of wind action and convection currents; (ii) the hypolimnion, a bottom layer of cold and relatively undisturbed water, which is isolated from wind and thermal effects and is generally devoid of light and dissolved oxygen, separated from the epilimnion by, (iii) the metalimnion, and intermediate region in which the temperature gradient, the thermocline, is steepest. (i)
The term thermocline has been defined variously, but correctly refers to the plane or surface of maximum rate of decrease of temperature with depth (Wetzel, 1975; Chiswell and Rauchle, 1986). The metalimnion in terms of an oxycline rather than thermocline has been used in relation to destratification and speciation of iron and manganese. The oxycline, over which there is a sudden change in dissolved oxygen (D.O.) values, may often coincide with the thermocline but is usually much more clearly cut. There are many different forms of stratification brought about by the form, size, depth and location of the lake basin, volume of through-flow, and effects of climate (Coker et al., 1979). With the onset of thermal stratification, deep circulation within the water column ceases and the oxygen transport from the productive zones near the surface to the deeper part of the water column is restricted. Bacterial consumption of oxygen does not cease following stratification, and in the absence of replacement from the surface zone the water/sediment interface rapidly becomes anaerobic and the overall oxygen content of the hypolimnion drops rapidly. The anaerobic production of carbon dioxide (Hutchinson, 1975), in conjunction with the lack of photosynthetic activity within the oxygen deficient region, usually results in the lowering of pH. Decomposition of settling organic material within the anoxic hypolimnion and at the water/sediment interface also leads to the release of hydrogen sulfide and ammonia. The development of chemically reducing conditions at the water/sediment interface leads to the release of iron and manganese compounds from the sediments into the water body (Uhlmann, 1979; Davison and Woof, 1984). Dissolved iron and manganese may adsorb onto biogenic and organically coated particles (Ballstrieri et al., 1981; Martin and Knauer, 1985), or oxidize during the formation of oxyhydroxide coatings on bacterial capsules (Cowen and Silver, 1984; Cowen and Bruland, 1985). Iron and manganese may be incorporated into organic tissue
LAKE
DESTRATIFICATION
AND
SPECIATION
OF FE AND
MN
435
as necessary biochemical micronutrients, perhaps to be redissolved upon the recycling of these m~iterials with depth. Thus biological stratification can also influence the distribution of iron and manganese in a water body. ARTIFICIAL DESTRATIFICATIONOF DAM WATER Aeration/destratification is a management technique which has been widely used internationally and in Australia to effect beneficial changes in the thermal, chemical and biochemical status of both man-made and naturally occurring water storage. The method is now recognized as an effective way of minimizing the release of iron and manganese from sediments and is being used increasingly in Australian reservoirs for water quality improvement purposes (Loos 1987). Oxygen saturation maintained at above 5% throughout the water column is sufficient to prevent high concentrations of iron and manganese from developing (Bowles et al. 1979). Two additional benefits of the introduction of artificial destratification are the maintenance of uniform water quality within the reservoir throughout the year, and the maintenance of maximum residence time of soluble metal ions within the reservoir. This must be done to ensure complete formation of particulate forms of iron and manganese with subsequent sedimentation of iron and manganese oxides from the water column (Waite et al., 1989).
Aims of the Research The work described here was undertaken with a view to developing a deeper understanding of the forms of occurence (speciation) of manganese and iron in impounded waters such as storage dams. Within this deeper understanding, there was seen the need to elucidate how speciation of manganese and iron changed, and what changes in naturally occuring parameters led to such changes. The work had a particular applied aim in seeking to use the knowledge obtained about speciation to develop both the understanding, and the application, of methods used to remove these nuisance metals from potable water during water treatment. Because studies were also carried out on an artifically destratified dam, new insights into the effectiveness of aeration in dealing with manganese and iron removal were also able to be studied.
Methodology Employed Two major techniques of manganese and iron speciation were developed for the first time for use on darn water storages. These were the use of electron paramagnetic spectroscopy (EPRS) to speciate dissolved manganese, and the use of secondary ion mass spectroscopy (SIMS) and electron spectroscopy for chemical analysis (ESCA) to speciate particulate manganese and iron. Other techniques used have previously been employed in a similar manner to the use made in the work here described.
436
BARRY CHISWELL AND MYINT ZAW
DISSOLVED AND PARTICULATESPECIES For the purposes of this work, the distinction between these two forms of a metal is related to passage through (dissolved) and retention by (particulate) a filter membrane; usually of pore size 0.45 #m, but sometimes 0.01 #m. SPECIATION OF DISSOLVEDMN AND FE
EPR Spectroscopy of Dissolved Mn This work has developed the use of EPRS to yield the analytical determination of aquated manganese(II) in freshwaters. The technique has been shown to be very sensitive (detection limit of 5 #g 1-1)and to be as accurate as flame atomic adsorption spectroscopy for manganese; it also has the great advantage of being selective for aquated manganese(II) only.
Gel Filtration~Molecular Weight Speciation of Mn and Fe The method of gel (or particle size exclusion) filtration has been applied to the waters from both North Pine and Hinze dams. The object of this work is to speciate manganese and iron dissolved species by molecular weight ranges. The gels used, Sephadex, G-10 and G-25, have approximate molecular weight operational ranges of 1 to 700 and 1 to 5000 Daltons respectively; in theory compounds within these molecular weight ranges are held on the chromatography column, while compounds with molecular weights above the upper limit of the range are excluded from the gel and pass out of the column. This work has been coupled with the use of the ion-exchange column material, Chelex100, which is a strong binder of cationic species, but allows the passage of neutral organic complexes.
Membrane Filtration and Speciation of Mn and Fe The use of both 0.45 #m and 0.01 #m membrane filters upon like samples has been applied to give further evidence upon molecular/particle size of species in samples.
Speciation of Dissolved Fe(II) and Fe(II1) This has been achieved by a variation of the colorimetric analytical method which uses the dye ferrozine. This material, which is ion-specific for Fe(II), can be used to first determine Fe(II) and then, after a reduction step upon the sample, can also determine the Fe(III) originally present, this analysis has been used in particular to estimate the levels of complexation of Fe(II) and Fe(III), by coupling this work with gel filtration results. SPECIATION OF PARTICULATE
MN AND FE
Surface and Bulk Analysis of Material by SIMS and ESCA Coupled with a detailed study of particulates, taken from different depths of a dam and during different seasons, by scanning electron microscopy (SEM), surface analysis techniques have been used to analyse the oxidation state and binding atoms for Mn and Fe
LAKE
DESTRATIFICATION
AND
SPECIATION
OF
FE
AND
MN
437
present both on the surface and in the bulk of the particulates. The technique, although not previously applied to such a study, yields excellent information upon the nature of the particulate surface, and then, by ion-bombardment to remove approximately the top ten atomic layers of the solid, the nature of the bulk surface can be studied.
Sequential Extraction and Differential Solubility Studies of Mn and Fe in Particulates The nature of binding of both metals in particulate samples has been studied by these two closely related techniques. Sequential extraction of samples consists of step-wise extraction using increasingly more vigorous reactants; a typical program would have the sequence: (i) (ii) (iii) (iv) (iv)
equilibration with magnesium chloride sodium acetate/acetic acid extraction hydroxylamine/acetic acid extraction hydrogen peroxide/acetic acid extraction hydrofluoric/perchloric acids extraction in an attempt to define the amounts of Mn and Fe present as: (i) absorbed ionic material, (ii) carbonate-bound metals, (iii) oxide-bound metals, (iv) organically-bound metals, (v) residual (probably in clays) metals, respectively.
Differential solubility in various acids has been coupled with this work. DAM WATERSSTUDIEDAND SAMPLINGPROGRAMSUSED Two dams in South East Queensland have been extensively studied by most of the methods mentioned above. These are North Pine dam, approximately 50 km north west of Brisbane, and Hinze dam, approximately 100 km south west of Brisbane. Both dams are major storages; North Pine has a capacity of 202 000 M1 and supplies the northern suburbs of Brisbane, while Hinze dam has a capacity of 164 000 M1 and serves most of the rapidly growing coastal area south of Brisbane. A third, small dam Lake Kurwongbah, which is very close to North Pine dam has also been studied in far less detail. Work since 1986 has involved regular (usually fortnightly - but sometimes weekly) sampling of the dams by research students and the leader of the Water Research Group. Such sampling has often been coupled with sampling programs undertaken by the local Council authorities, however, it has been a rule of sampling that each worker collects and pre-treats his/her own samples. Sampling for depth samples was undertaken by use of double-ended inlet-outlet depth sampler of various (including a 'home-made' version constructed in the Chemistry department workshop) makes, while sediment samples were obtained by a grab sampler which penetrated approximately 5 cm into the sediment. Attempts to sample top layers of the sediment were not successful; an electronic eye system which supposedly measured
438
BARRY
CHISWELL
AND
MYINT
ZAW
when a micro-sampler was entering the sediment, worked within a few metres of the water surface. In all three dams the algal population is always such that one cannot see the bottom of the dam at even a depth of a metre. Samples once collected were stored in polythene bottles, which had been thoroughly cleaned by repeated acid washing followed by distilled water cleansing. Filtration of samples was often done immediately following sampling, although this was not possible when using 0.01/~m membranes. Parameters such as pH, dissolved oxygen, temperature and conductivity where either measured in situ or immediately following collection of the sample. Results and Discussion
SPECIATION OF DISSOLVEDMANGANESE The method of electron paramagnetic resonance spectroscopy (EPRS) (Chiswell and Mokhtar, 1986 and 1987) has been employed upon waters obtained from the epilimnion, the hypolimnion and the sediment of the three different dams to show: (i) that waters taken from any depth in the dam, and filtered through a 0.01 #m membrane contain dissolved manganese in the form of aquated Mn(II) only; such filtrate samples have identical concentrations of Mn(II) aq as determined by EPRS and atomic absorption spectroscopy (AAS) (see Table I). (ii) that waters filtered through a 0.45 ~t membrane from the epilimnic regions of the dam may contain a small concentration (
